The allocation of cells into the three early embryonic germ layers, in vivo, follows a specific patio-temporal sequence. In response to growth factors, human embryonic stem cells (hESCs) can generate the three germ layers in culture, but differentiation is random and spatially disordered. We have demonstrated that when hESCs are grown on micro-patterned surfaces in colonies of defined size and shape, they differentiate with quantitatively reproducible spatial patterns of gene expression suggestive of those in mammalian embryos. This is the first demonstration of ordered germ layer patterning in vitro and opens the door to investigations of early development currently impossible to perform in intact mammalian embryos, particularly human embryos. Since the spatial pattern develops spontaneously in a well-mixed medium, this system affords a quantitative phenotype with which to analyze cell-cell signaling. The goal of this grant is to establish micro patterned cell cultureas a system for studying signaling dynamics and spatial patterning in hESCs. We propose to further validate our assay by differentiating mouse epiblast stem cells (the closest analogue to hESC) on our substrates and comparing with mouse embryos purchased from commercial sources. We will stimulate the TGF-?, BMP, and Wnt pathways that pattern the early mammalian embryo and follow the transcriptional effectors for these pathways as well as the resulting pattern of germ layer markers in time. We will engineer cell lines with combinations of fluorescent reporters for signaling and fate and use these for time-lapse imaging. The results will be fit to a phenotypic computational model that will permit us to rationally design combinations of ligands with defined doses and timings that produce a defined outcome. We will also manipulate the levels of secreted ligands (both activators and inhibitors) that are responsible for paracrine signaling in the embryo on our micro patterned colonies and further quantify cell-cell interactions. Physical assays for secreted ligands using microfluidics will be employed to quantify mechanisms of ligand transport.